There is compelling evidence that carcinogenesis is a multistep process and multiple genetic lesions are necessary to develop cancer in human. Along the genetic lesions, the alteration of p53 protein (due to mutation of p53 gene or by infection of "high risk" human papillomaviruses [HPV], e.g., type 16 or 18 HPV) is the most frequently found genetic disorder in human cancers including oral cancer. A significant body of evidence supports p53 mutation as an early event in oral carcinogenesis. Our studies have also shown that human oral keratinocytes (HOK) containing negligible amount of wild-type (wt) p53 protein (because of HPV DNA integration) and HOK expressing mutant (mt) p53 protein are immortal, but not tumorigenic. These cells convert to tumorigenic cells when exposed to tobacco-carcinogens. whereas cells with a normal complement of p53 do not, indicating that p53 dysfunction appears to be an early event in oral carcinogenesis. Therefore, the dysfunction of p53 protein appears be an early event at least in oral carcinogenesis and also be necessary for subsequent genetic disorders of other genes to convert normal cells to tumor cells in the human oral cavity. Inasmuch as wt p53 protein plays a major role in the regulation of cell cycle arrest, we hypothesize that normal human oral keratinocytes containing wt p53 protein repair damaged DNA more efficiently than oral keratinocytes with defective p53 function. As demonstrated by many studies including our preliminary data, cells expressing wt p53 protein have the ability to establish transient delays in the progression of cell cycle when exposed to genotoxic agents, but cells with defective p53 function do not possess such ability. Since the transient arrest of the cell cycle progression is necessary to repair damaged DNA prior to replication of damaged DNA template and segregation of damaged chromosome, cells with defective p53 function may fail or have limited ability to repair the damaged DNA when exposed to genotoxic agents. In the proposed study, we will test the above hypothesis by determining the effect of major tobacco- carcinogens on (1) the progression of cell cycle, the expression of major growth arrest and DNA damage inducible genes (e.g., p53, WAF1/CIP1, and gadd45), and the activity of cyclin-dependent kinases (cdks); (2) the genotoxicity of host chromosomal DNA (DNA adducts and single strand DNA breaks); (3) the repair of damaged DNA; and (4) the mutation frequencies and spectrum in normal HOK expressing wt p53, HOK expressing HPV-16 or HPV-18 E6 protein, HOK expressing mutant p53 protein, and HPV-immortalized oral keratinocytes. These proposed studies should help us gain more insight into molecular mechanisms of tobacco-related oral carcinogenesis.